Large Language Model (LLM) unlearning is essential for removing sensitive or copyrighted knowledge while preserving general utility. Existing methods often leave residual knowledge in intermediate representations, which can still be recovered. To address this, we propose Cascade, a hierarchical recoverability control framework that minimizes the internal identifiability of target knowledge. Cascade combines three complementary controls: path-level routing to suppress privacy-associated activation routes, representation-level compression to reduce geometric separability, and decoding-level intervention to limit residual recovery. Experiments on TOFU, MUSE-News, and WMDP, including robustness tests with query reformulation and extraction-style prompts, show that Cascade effectively reduces recoverability while maintaining stable model utility.

Rubric-based rewards are increasingly used for open-ended language model post-training, but criterion-level scores are commonly aggregated as independent utilities. This flat scalarization ignores semantic dependencies among criteria, allowing downstream credit or penalties to be counted even when the conditions that license them are absent. We call this structural failure False Credit Propagation (FCP).

We propose GEAR (Graphical Event Aggregation for Rubric rewards), a dependency-aware aggregation framework for rubric rewards. GEAR represents rubric criteria as latent events in a typed dependency graph, uses monotone retention factors to infer dependency-adjusted criterion marginals, and computes a normalized expected-utility reward.

This preserves compatibility with standard rubric-based reinforcement learning while adding only linear-time aggregation over criteria and edges. Experiments on HealthBench, WritingBench, and PLawBench with two policy backbones show that GEAR consistently improves over flat aggregation and hard gating, both in controlled aggregation-only settings and as a plug-in module for existing rubric-guided RL pipelines.

FCP diagnostics further show that GEAR substantially reduces invalid downstream utility while avoiding the over-suppression of deterministic gating.

LLM agents are increasingly expected to operate across heterogeneous task regimes that require distinct execution paradigms. This challenges fixed agent systems and motivates system-level meta-adaptation beyond isolated component updates. While existing works have adapted external harness or trained underlying reasoning policies, full-system adaptation remains insufficiently characterized. The adaptation space between structure and execution is rarely made explicit, and the compatibility between the external harness and the internal reasoner is not optimized jointly. We propose HarnessForge, a meta-adaptive framework for evolving LLM agent systems. HarnessForge formulates an agent system as a harness–policy pair, defining a stable adaptation space that separates harness-level execution structure from policy-level reasoning behavior. It then performs harness–policy co-evolution through fault-guided harness tailoring and harness-conditioned policy alignment. Experiments across five benchmarks from diverse domains show that HarnessForge consistently improves both Qwen3-4B and Qwen3-8B backbones, outperforming harness-only and policy-only baselines with gains of up to 12.0% over the strongest baseline and achieving favorable rollout-efficiency tradeoffs, demonstrating that harness–policy co-evolution is effective, and that executable compatibility between the harness and reasoning policy is essential for agent-system adaptation.

In code vulnerability detection tasks, a detector trained on a label-rich source domain fails to provide accurate prediction on new or unseen target domains due to the lack of labeled training data on target domains. Previous studies mainly utilize domain adaptation to perform cross-domain vulnerability detection. But they ignore the negative effect of private semantic characteristics of the target domain for domain alignment, which easily causes the problem of negative transfer. In addition, these methods forcibly reduce the distribution discrepancy between domains and do not take into account the interference of irrelevant target instances for distributional domain alignment, which leads to the problem of excessive alignment. To address the above issues, we propose a novel cross-domain code vulnerability detection framework named MNCRI. Specifically, we introduce mutual nearest neighbor contrastive learning to align the source domain and target domain geometrically, which could align the common semantic characteristics of two domains and separate out the private semantic characteristics of each domain. Furthermore, we introduce an instance re-weighting scheme to alleviate the problem of excessive alignment. This scheme dynamically assign different weights to instances, reducing the contribution of irrelevant instances so as to achieve better domain alignment. Finally, extensive experiments demonstrate that MNCRI significantly outperforms state-of-the-art cross-domain code vulnerability detection methods by a large margin.